Towards functional biochips: a new approach to pattern
proteins on nano carriers
Jan Pille
Supervisors: J.C.M. van Hest and T. Michon
Institute for Molecules and Materials, Radboud University Nijmegen, The Netherlands
Biologie du Fruit et Pathologie, INRA, Villenave d’Ornon, France
contact: [email protected]
ABSTRACT
Patterning fluorescent proteins and enzymes on the
surface of well-defined scaffolds made from biological
nano carriers is a promising tool in the creation of
biological sensors based on rational design. Here we
develop a new, versatile block-building method to
address the critical step in the fabrication of such
biological devices; the topologically controlled coupling
of proteins to their macromolecular scaffolds. We utilize
a fluorescent protein that is genetically fused to an
antibody-binding peptide to specifically coat plant virus
particles. We validate our approach by affinity assays and
correlative microscopy.
Keywords
Biochips, biosensors, correlative microscopy, monomeric
yellow fluorescent protein, nano carriers, supramolecular
assembly, zucchini yellow mosaic virus, Z33.
INTRODUCTION
Patterning fluorescent proteins and enzymes on
macromolecular scaffolds is a promising technique for the
generation of biosensors or -chips. Nano carriers (NCs)
such as viruses1 can be arranged on supports using dippen nanolithography2. These chips can then be further
utilized by immobilizing proteins on their surface. The
critical feature during the fabrication of such chips is the
linking process between the protein of interest and the
scaffold. The coupling method used should ensure a high
and well-defined coverage while avoiding adverse effects
on the scaffolds stability and protein activity. So far, a
variety of approaches have been developed; for example,
transferrin was covalently bound to the surface of
icosahedra viruses. First, unnatural amino acids were
incorporated in the virus. Transferring was then bound by
copper-catalyzed azide-alkyne cycloaddition3. This
strategy, while successful for two icosahedral viruses,
suffers from poor versatility. Changing the peptide
sequence of viral proteins can be deleterious for their
assembly. Furthermore, subsequent chemical reactions to
achieve covalent coupling may have adverse effects on
virus stability. Genetic fusion of proteins and enzymes to
viral proteins has been successful to some extent, but
fusion proteins can prevent virus assembly due to steric
hindrance4. Additionally, these constructs suffer from
poor stability, resulting in rapid exclusion of inserted
genes5. Expressed proteins may also show toxicity in the
host organism6. Therefore, non covalent, high affinity
patterning of proteins seems desirable to avoid these
disadvantages. Indeed, charged coiled-coil motifs that are
able to dimerize, so-called leucine zippers, have been
used to entrap fluorescent proteins in the interior of
cowpea chlorotic mottle virus7. Still, the virus had to be
altered genetically and assembly was effective only in
vitro in the presence of wild-type coat proteins. This
approach is not feasible in the case of larger viruses that
do not assemble in vitro, but which are more suitable for
forming supramolecular chip-like scaffolds. Here we
develop a new, non-covalent and versatile approach that
utilizes native viruses. It is based on three different
building blocks, which can be assembled very efficiently:
Zucchini yellow mosaic virus, α-ZYMV immunoglobulin
G and the IgG-binding peptide Z33.
Zucchini Yellow Mosaic Virus
Zucchini yellow mosaic virus (ZYMV) is a filamentous
potyvirus consisting of about 2000 coat proteins (CPs),
with a size of 36 kDa each. They are regularly assembled
in a helical fashion around the genetic material. ZYMV
has a length of ~700 nm and a diameter of 11 nm.
Therefore, it offers a nano-sized regular surface that can
be functionalized while being suitable for chip-like
arrangements on supports. ZYMV is stable under a wide
range of pH and temperatures and is easily produced in
fast-growing plant species8.
α-ZYMV immunoglobulin G
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. By sending in this
paper the student gives permission to post this paper on the VSNU website
and digital student journals bearing this notice and the full citation on the
first page.
SRC 2013, November 20, 2013, Amsterdam, The Netherlands.
Copyright 2013 SRC / VSNU
Immunoglobulins , or antibodies, have an unmet affinity
and specificity against their target due to their variable
domains, the so-called Fab region. They also offer a
constant domain, the Fc region, that depends of the class
of antibody. Immunoglobulin G (IgG) is an antibody
isotope with a size of typically 150 kDa (two heavy
chains of 50 kDa and two light chains of 25 kDa each).
Antibodies may serve as an ideal mediator of
supramolecular assembly, as IgGs raised against ZYMV
are able to coat its surface selectively and efficiently
while retaining a defined surface that can further be
utilized.
The peptide Z33
The peptide Z33 stems from the staphylococcal protein A
(SPA)9. SPA recognizes a broad range of IgGs from
various species. Due to its bulky size of 56kDa and its
membrane-bound nature, which makes it impractical to
utilize directly, efforts have been taken to minimize its
binding domains. The smallest peptide derived, consisting
of only 33 amino acids, is known as Z33 and is capable of
binding the Fc region of IgGs with a high affinity10. Z33
has been utilized to target adenovirus vectors11 and vault
nanoparticles12 specifically to the surface of cancer cells,
but it has never been used for controlled, supramolecular
assembly of functional materials.
In this study, Z33 was fused genetically to the monomeric
yellow fluorescent protein (mYFP)13, yielding Z33mYFP
with a size of 33 kDa. The supramolecular assembly of
ZYMV, IgGs and Z33mYFP (Figure 1) was
subsequently studied. We used affinity experiments and
correlative microscopy to show that this approach serves
as a robust, efficient and versatile block-building method
to pattern proteins on the surface of macromolecular nano
carriers while retaining protein activity.
Figure 1: Illustration of
the
approach.
The
different building blocks
are ZYMV particles,
IgGs raised against the
coat proteins of ZYMV
and Z33mYFP. ZYMV
serves as a scaffold for
the antibody. The Fc
region of the IgG enables
binding of the peptide
Z33, which is genetically
fused to mYFP. Figure
adapted from Pille et al14.
MATERIALS & METHODS
All chemicals and reagents used were purchased from
Sigma-Aldrich unless stated otherwise. Polymerase chain
reactions (PCR) were carried out on a BIO-RAD
PTC0200 DNA Engine Peltier Thermal Cycler. For
transmission electron microscopy (TEM), a CM10 FEI
was used. Fluorescence microscopy was performed on an
epifluorescence E800 Nikon microscope with an HQ2
CoolSNAP CCD detector. Purified ZYMV (E15 strain)
and rabbit-born α-ZYMV IgGs were kindly supplied by
H. Lecoq from INRA (Montfavet).
Cloning and expression of Z33mYFP
The sequence of Z33 (peptide sequence: FNMQQQRR
FYEALHDPNLNEEQRNAKIKSIRDD10) was fused to
the amino terminus of mYFP, including a bridging
sequence encoding GGGGS to ensure flexibility of the
Z33 peptide. It was transformed into the pET21a(+)
(Novagen) expression vector by standard techniques of
recombinant cloning (data not shown). Z33mYFP was
overexpressed in E.coli BL21(DE3)pLysS (Novagen)
cells and purified by ion metal affinity chromatography
(IMAC) with a yield of around 160 mg per liter culture.
Purity and size was validated by sodium dodecyl sulfate
polyacrylamide gel electrophoresis (SDS-PAGE) and
matrix-assisted laser desorption ionization - time of flight
(data not shown).
Affinity assay
Antibodies used for affinity assays were rabbit-born
polyclonal IgGs raised against whole ZYMV particles.
Antibodies were mixed with Z33mYFP (molar ratio 1:1
and 1:5) and allowed to incubate 5-20 min at room
temperature in phosphate buffered saline (PBS) pH 7.4.
Following incubation, Z33mYFP was repeatedly purified
by IMAC and washed with PBS pH 7.4 before being
loaded on a 12% SDS-PAGE gel; staining was done with
InstantBlue (Expedeon).
Correlative microscopy
For correlative microscopy, a carbon-coated copper
finder-grid was prepared according to the following
scheme:
Step
Applied Protein
(concentration)
Incubation
time (min)
1
2
3
4
ZYMV (0,06mg/mL)
BSA (0,1% w/v)
α-ZYMV (0,02 mg/mL)
Z33mYFP (0,02mg/mL)
5
60
60
60
Rabbit α-ZYMV polyclonal IgGs and Z33mYFP were
diluted in 'preparation solution' (filtered 0,1 M sodium
phosphate buffer (SPB) pH 8 containing 0,1 % w/v
bovine serum albumin (BSA) and 0,1 % v/v Tween 20).
ZYMV particles where diluted in SPB pH 8. After each
step, the grids were washed twice with preparation
solution. After preparation, the grid was placed on a glass
slide. One drop of filtered SPB pH 8 was placed on the
grid to ensure a thin water layer between the two glass
slides and tape was used to keep a distance of ~1,5 mm
between slides. Fluorescence was measured between 510
- 560 nm after excitation with wavelengths of 460 - 500
nm. Subsequently, the grid was dried on paper, stained
with 2% phosphotungstic acid and observed via
transmission electron microscopy.
Supramolecular assembly in solution
ZYMV particles were mixed with α-ZYMV antibodies
and Z33mYFP in a 1:1:5 molar ratio (for ZYMV, the
molar ratio refers to the number of CPs present) in SPB
pH 8 and were incubated for 1h at room temperature. The
sample was dialyzed against a 100 times larger volume of
SPB pH 8 with 12 buffer changes over the course of 96h
at 4°C in microdialysis knobs of 50μL attached to a
cellulose ester dialysis membrane (Spectra/Por Biotech)
with a MWCO of 300kDa. Samples missing either virus
particles or antibody were used to control for the
diffusion of unbound proteins. Subsequently, the samples
were loaded on a 12% SDS-PAGE gel and stained with
InstantBlue (Expedeon). Known quantities of virus,
antibody and Z33mYFP were used to create calibration
curves to determine the amount of the different proteins
present. Quantitative analysis was done by densitometry
(ImageJ 1.45S).
RESULTS & DISCUSSION
Affinity assay
To determine whether Z33mYFP is capable of binding
antibodies, an affinity assay was used. Z33mYFP and αZYMV were incubated and Z33mYFP was subsequently
purified by IMAC. Figure 2 shows the results of the
affinity assay. When no Z33mYFP was present, α-ZYMV
IgG was most prominent in the first wash (left gel, W1)
as expected, and got diluted in subsequent cycles (left gel,
W2-3); almost no antibody was present in the final
elution (left gel, E). When Z33mYFP and α-ZYMV were
mixed in a 1:1 molar ratio, Z33mYFP seemed capable of
binding a fraction of the present antibodies, as the
majority of IgGs were visible in the elution phase
together with Z33mYFP (middle gel, E). The Z33 peptide
has two possible binding sites on one antibody (one on
each heavy chain), a 1:1 molar ratio was expected to only
partially bind the present antibodies. When Z33mYFP
was present in fivefold excess, no antibody was visible in
the washing phases, but was fully recovered in the elution
phase together with Z33mYFP (right gel, E). This assay
proves that the Z33 peptide is still capable of binding αZYMV IgGs with high affinity even when fused to
mYFP. Subsequently, ZYMV particles were coated with
antibody and fluorescent Z33mYFP and visualized by
correlative microscopy.
Figure 2: SDS-PAGE gel of affinity assay. Left gel: α-ZYMV
antibodies alone were subsequently diluted through the washing
steps (W1-W3); almost no antibody was left in the elution phase
(E). Middle gel: When Z33mYFP and α-ZYMV antibody were
mixed in a 1:1 molar ratio, Z33mYFP eluted together with a
large fraction of antibodies. Right gel: When Z33mYFP was
present in fivefold excess, no antibody was visible in the
different washing phases, but was fully recovered in the elution
phase together with Z33mYFP. Abbreviations: Ab, antibody; hc,
heavy chain; lc, light chain. Figure adapted from Pille et al14.
Correlative microscopy
Figure 3 shows pictures of the same spot on a grid
observed with either epifluorescence (Figure 3a, c) or by
transmission electron microscopy (Figure 3b, d and e).
In the depicted areas, the same pattern connecting either
fluorescent 'spots' or virus particles can be drawn. The
orientation, intensity and size of the fluorescent spots
correspond with the virus particles observed by TEM.
This indicates that the virus particles are coated
efficiently with fluorescent Z33mYFP molecules and
shows that fluorescent activity was retained. To
investigate the coating efficiency, the complex was
assembled in solution and analyzed after dialysis.
Supramolecular assembly in solution
Figure 4 shows the results of the supramolecular
assembly in solution. Only when α-ZYMV antibodies and
ZYMV were both present, Z33mYFP was retained in the
sample after dialysis. From the intensity of the respective
bands, the amount of protein present in the sample was
calculated by utilizing a calibration curve based on band
intensity of pure protein (data not shown). It was
calculated that the retention of Z33mYFP in respect to
ZYMV coat proteins is around 0,87:1, indicating a
coverage of approximately 87%. These results are in
agreement with the results from correlative microscopy,
as Z33mYFP-coated virus particles showed fluorescence
covering each corresponding virus completely, indicating
a high coupling efficiency. This shows that the formed
complex is stable for extended periods of time, as the
dialysis was carried out over the course of four days. The
theoretical maximum coverage would be 200%, as each
viral coat protein could be bound by one antibody bearing
two Z33mYFP molecules. To our knowledge, the
efficiency achieved exceeds all known supramolecular
NC systems that use affinity-based assembly techniques.
Figure 3: Correlative microscopy of ZYMV coated with αZYMV IgGs and Z33mYFP. The fluorescence emitted by
Z33mYFP clusters was first localized by epifluorescence (a, c)
before observing virus particles by transmission electron
microscopy (b, d, and e). The pattern of fluorescent spots
correspond with the size and orientation of virus particles (c, d).
A single virus particle (e) is shown for illustration. Figure
adapted from Pille et al14.
Figure 4: SDS-PAGE gel showing the different samples after
dialysis. Only when α-ZYMV antibodies (heavy chains depicted)
and ZYMV particles were both present (right lane), Z33mYFP
was retained in the sample. Whole ZYMV particles were
retained due to their large size. For 1 µg of ZYMV CPs, 0,79 µg
of Z33mYFP was retained, which means that for 1 mol of
ZYMV CPs, 0,87 mol of Z33mYFP was retained. Figure adapted
from Pille et al14.
CONCLUSION
We have shown the development of a new approach to
coat virus particles with a protein of interest. We fused
the antibody-binding peptide Z33 genetically to the
monomeric yellow fluorescent protein. Subsequently, we
patterned it on the macromolecular structure of zucchini
yellow mosaic virus utilizing antibodies as mediators of
assembly. We validated our approach by affinity assays
and correlative spectroscopy. We showed that the activity
of the fluorescent protein was retained after assembly and
that the coverage efficiency was high (87%). This
approach does not depend on chemically or genetically
changing the virus particle and should therefore facilitate
the expansion to other macromolecular scaffolds. The
principle of using a scaffold-specific antibody and Z33
fusion protein can be extended to a wide variety of
proteins and enzymes to create functional biochips with
fluorescent properties and/or catalytic activity. We
recently showed that the enzyme 4-coumarate:CoA-ligase
2 can be functionalized with the Z33 peptide and that it
retains its catalytic activity after supramolecular
assembly14. Utilizing the developed approach, future
work will focus on patterning enzymes that catalyze
subsequent reactions on virus particles and to assembly
functional chips of viruses coated by fluorescent proteins
and enzymes.
ROLE OF THE STUDENT
J. Pille was an undergraduate student working under the
supervision of J.C.M. van Hest and T. Michon when the
research in the report was performed. The topic was
developed with contributions from J.C.M. van Hest, T.
Michon, F. Smits, M. van Eldijk, N. Carette, D. Cardinale
and J. Pille. Experiments, analysis and writing of the
report were done by J. Pille.
ACKNOWLEDGMENTS
We thank the Tsien laboratory (Howard Hughes Medical
Institute, University of California, San Diego) for
providing the mYFP gene. We thank the Bordeaux
Imaging Center (Centre de Genomique Fonctionnelle de
Bordeaux) for providing facilities, L. Brocard and B.
Batailler for instrumental training. This work was
founded by the Agence Nationale pour la Recherche,
France and by the Radboud Honours Academy, Radboud
University Nijmegen.
REFERENCES
1. Cardinale, D.; Carette, N.; Michon, T., Virus scaffolds
as enzyme nano-carriers. Trends Biotechnol 2012, 30 (7),
369-76.
2. Vega, R.; Maspoch, D.; Salaita, K.; Mirkin, C.,
Nanoarrays of single virus particles. Angewandte Chemie
(International ed. in English) 2005, 44 (37), 6013-6015.
3. Strable, E.; Prasuhn, D.; Udit, A.; Brown, S.; Link, A.;
Ngo, J.; Lander, G.; Quispe, J.; Potter, C.; Carragher, B.;
Tirrell, D.; Finn, M., Unnatural amino acid incorporation
into virus-like particles. Bioconjugate chemistry 2008, 19
(4), 866-875.
4. Werner, S.; Marillonnet, S.; Hause, G.; Klimyuk, V.;
Gleba, Y., Immunoabsorbent nanoparticles based on a
tobamovirus displaying protein A. P Natl Acad Sci USA
2006, 103 (47), 17678-17683.
5. (a) Beauchemin, C.; Bougie, V.; Laliberté, J.-F.,
Simultaneous production of two foreign proteins from a
polyvirus-based vector. Virus research 2005, 112 (1-2),
1-8; (b) Wang, Z. D.; Ueda, S.; Uyeda, I.; Yagihashi, H.;
Sekiguchi, H.; Tacahashi, Y.; Sato, M.; Ohya, K.;
Sugimoto, C.; Matsumura, T., Positional effect of gene
insertion on genetic stability of a clover yellow vein
virus-based expression vector. J Gen Plant Pathol 2003,
69 (5), 327-334.
6. Carette, N.; Engelkamp, H.; Akpa, E.; Pierre, S.;
Cameron, N.; Christianen, P.; Maan, J.; Thies, J.;
Weberskirch, R.; Rowan, A.; Nolte, R.; Michon, T.; Van
Hest,
J.,
A
virus-based
biocatalyst.
Nature
nanotechnology 2007, 2 (4), 226-229.
7. Minten, I.; Nolte, R.; Cornelissen, J., Complex
assembly behavior during the encapsulation of green
fluorescent protein analogs in virus derived protein
capsules. Macromolecular bioscience 2010, 10 (5), 539545.
8. Desbiez, C.; Lecoq, H., Zucchini yellow mosaic virus.
Plant Pathology 1997, 46 (6), 809-829.
9. Graille, M.; Stura, E. A.; Corper, A. L.; Sutton, B. J.;
Taussig, M. J.; Charbonnier, J. B.; Silverman, G. J.,
Crystal structure of a Staphylococcus aureus protein A
domain complexed with the Fab fragment of a human
IgM antibody: structural basis for recognition of B-cell
receptors and superantigen activity. Proc Natl Acad Sci U
S A 2000, 97 (10), 5399-404.
10. Braisted, A. C.; Wells, J. A., Minimizing a binding
domain from protein A. Proc Natl Acad Sci U S A 1996,
93 (12), 5688-92.
11. Volpers, C.; Thirion, C.; Biermann, V.; Hussmann,
S.; Kewes, H.; Dunant, P.; von der Mark, H.; Herrmann,
A.; Kochanek, S.; Lochmuller, H., Antibody-mediated
targeting of an adenovirus vector modified to contain a
synthetic immunoglobulin G-binding domain in the
capsid. Journal of Virology 2003, 77 (3), 2093-2104.
12. Kickhoefer, V. A.; Han, M.; Raval-Fernandes, S.;
Poderycki, M. J.; Moniz, R. J.; Vaccari, D.; Silvestry, M.;
Stewart, P. L.; Kelly, K. A.; Rome, L. H., Targeting vault
nanoparticles to specific cell surface receptors. ACS Nano
2009, 3 (1), 27-36.
13. Zacharias, D.; Violin, J.; Newton, A.; Tsien, R.,
Partitioning of lipid-modified monomeric GFPs into
membrane microdomains of live cells. Science (New
York, N.Y.) 2002, 296 (5569), 913-916.
14. Pille, J.; Cardinale, D.; Carette, N.; Di Primo, C.;
Besong-Ndika, J.; Walter, J.; Lecoq, H.; van Eldijk, M.
B.; Smits, F. C. M.; Schoffelen, S.; van Hest, J. C. M.;
Mäkinen, K.; Michon, T., A General Strategy for Ordered
Non-covalent Protein Assembly on Well-defined Nanoscaffolds. Biomacromolecules (under revision).